INDEX

Explanations

references to individuals with titles such as "Lord"

oai_token-act-pair · gpt-3.5-turbo

occurrences of the title "Lord" in various contexts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

¥ŀ

-0.84

ickr

-0.76

urses

-0.74

Accessory

-0.69

OPLE

-0.69

misc

-0.68

single

-0.66

insula

-0.64

 proble

-0.64

ascus

-0.64

POSITIVE LOGITS

lord

1.03

Lord

0.93

 Vader

0.89

ess

0.88

stown

0.85

 Jesus

0.84

esses

0.82

uin

0.80

frey

0.80

 Lord

0.80

Activations Density 0.008%

references to individuals with titles such as "Lord"

occurrences of the title "Lord" in various contexts

No Comments

No Known Activations